In Brazil, routine verification of alcohol use amongst drivers is performed using breath alcohol analyzers and confirmation of ethanol in blood, using the headspace (HS) technique associated with gas chromatography with a flame ionization detector (GC/FID). Oral fluid (OF) is an alternative that, once collected, can be used both for screening and confirmation, and has many advantages. We propose an optimization of the extraction of ethanol from OF using the HS through experimental design and subsequent development and validation of an analytical method using HS-GC/FID and HS-GC/MS (mass spectrometry), using Quantisal® as a collection device. Experimental design was performed using the Box–Behnken design, and the evaluated parameters were heating temperature, stirring time and injection volume. Selectivity, residual effect, matrix effect, linearity, precision, accuracy, limits of detection and quantification, stability and recovery were evaluated in the validation process. The optimized conditions for extraction were: a temperature of 90 °C, an injection volume of 1000 μL, and a stirring time of 7 min. Linearity was obtained with an R2 greater than 0.99. The accuracy for the quality control samples remained within 101.56 and 111.2 of the target concentrations, whilst the precision did not exceed 12% of their relative standard deviations. The developed method showed full viability of performance, proving to be rapid and sensitive, as it does not require sample preparation steps. The HS-GC/MS method had detection limits lower than those of HS-GC/FID, and can be easily applied for routine confirmation of ethanol in drivers.
